This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

Transcription

1 This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier s archiving and manuscript policies are encouraged to visit:

2 Journal of Magnetic Resonance 202 (2010) 57 6 Contents lists available at ScienceDirect Journal of Magnetic Resonance journal homepage: Line-narrowing in proton-detected nitrogen-14 NMR Simone Cavadini a, Veronika Vitzthum a, Simone Ulzega a, *, Anuji Abraham a,1, Geoffrey Bodenhausen a,b a Institut des Sciences et Ingénierie Chimiques, Ecole Polytechnique Fédérale de Lausanne, Batochime 150, C-1015 Lausanne, Switzerland b Département de Chimie, associé au CNRS, Ecole Normale Supérieure, 24 rue Lhomond 7521, Paris Cedex 05, France article info abstract Article history: Received 14 August 2009 Revised 28 September 2009 Available online 0 September 2009 Keywords: Solid-state NMR Magic angle spinning (MAS) 14 N MQC SQC omonuclear dipolar decoupling Symmetry-based pulse sequence In solids spinning at the magic angle, the indirect detection of single-quantum (SQ) and double-quantum (DQ) 14 N spectra (I = 1) via spy nuclei S = 1/2 such as protons can be achieved in the manner of heteronuclear single- or multiple-quantum correlation (SQC or MQC) spectroscopy. The MQC method relies on the excitation of two-spin coherences of the type T I 11 TS 11 and TI 21 TS 11 at the beginning of the evolution interval t 1. The spectra obtained by Fourier transformation from t 1 to x 1 may be broadened by the homogenous decay of the transverse terms of the spy nuclei S. This broadening is mostly due to homonuclear dipolar S S 0 interactions between the proton spy nuclei. In this work we have investigated the possibility of inserting rotor-synchronized symmetry-based C or R sequences and decoupling schemes such as Phase-Modulated Lee Goldburg (PMLG) sequences in the evolution period. These schemes reduce the homonuclear proton proton interactions and lead to an enhancement of the resolution of both SQ and DQ proton-detected 14 N MQC spectra. In addition, we have investigated the combination of SQC with symmetry-based sequences and PMLG and shown that the highest resolution in the 14 N dimension is achieved by using SQC in combination with symmetry-based sequences of the R-type. We show improvements in resolution in samples of L-alanine and the tripeptide ala-ala-gly (AAG). In particular, for L-alanine the width of the 14 N SQ peak is reduced from 2 to 1.2 kz, in agreement with simulations. We report accurate measurements of quadrupolar coupling constants and asymmetry parameters for amide 14 N in AAG peptide bonds. Ó 2009 Elsevier Inc. All rights reserved. 1. Introduction Because of the ubiquity of nitrogen in a wide range of materials, and because of its structural and functional role in many proteins and nucleic acids, it would be attractive to obtain spectroscopic signatures of 14 N in a straightforward manner. In contrast to popular nuclei with S = 1/2 such as 1 C and 15 N, 14 N has a spin quantum number I = 1 and a challenging quadrupole coupling constant on the order of a few Mz [1 4]. In spite of these adverse conditions, 14 N spectroscopy appears to be on the verge of becoming a routine technique, thanks to indirect detection via neighboring spy nuclei with S = 1/2 such as 1 Cor 1 combined with magic angle spinning (MAS). In heteronuclear multiple-quantum correlation (MQC) [5 14], it is possible to transfer coherence from S to I nuclei via second-order quadrupole dipole cross-terms [15,16], also known as residual dipolar splittings (RDS) [9,11], via scalar couplings [10,14], or via recoupled heteronuclear dipolar interactions by using rotary resonance [14,17 20]. Recoupling can be achieved by applying a continuous radiofrequency (rf) field to the * Corresponding author. Fax: address: simone.ulzega@epfl.ch (S. Ulzega). 1 Present address: Department of Chemistry, Durham University, Durham, UK. spy nucleus S = 1/2 with an rf amplitude that is a multiple of the rotor spinning frequency, i.e., x S 1 = nx rot with n = 1 or 2. Alternatively, symmetry-based RN sequences of the R type [21,22] can recouple the heteronuclear dipolar I S ( 14 N 1 ) interactions while simultaneously decoupling the homonuclear S S 0 interactions [2]. It was found by van Beek et al. [24] that the R method was also efficient to recouple heteronuclear dipolar 17 O 1 interactions, and Brinkmann et al. [25,26] showed that it is possible to obtain accurate measurements of 17 O 1 distances by supercycling the shorter R4 1 sequence. Proton-detected MQC 2 experiments suffer from line-broadening due to homonuclear dipolar S S 0 ( 1 1 ) interactions in both x 1 and x 2 dimensions. The quest for high-resolution proton spectra in solid-state NMR has lead to the development of sophisticated methods for decoupling homonuclear dipolar 1 1 interactions. Recent progress in the design of MAS probes for very fast spinning, currently up to about 70 kz, makes it possible to attenuate these homonuclear interactions significantly [27]. The combination of fast MAS with multiple pulse sequences, known as combined rotation and multiple pulse spectroscopy (CRAMPS) [28], requires synchronization with the rotor period s rot =1/m rot, as in modified W4 sequences [29] or symmetry-based R sequences [0,1]. Synchronization is not a particular problem in the present context, since the t /$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi: /j.jmr

3 58 S. Cavadini et al. / Journal of Magnetic Resonance 202 (2010) 57 6 increments have to be rotor-synchronized anyway to eliminate the large first-order 14 N quadrupolar broadening. On the other hand, one can chose methods that do not require any synchronization, such as Lee Goldburg (LG) experiments [2] from which frequency-switched LG (FSLG) [,4] and Phase-Modulated Lee Goldburg (PMLG) [5 9] techniques have been derived. An alternative to these methods is offered by the DUMBO decoupling scheme [40], which leads to efficient decoupling over a wide range of spinning speeds [41,42]. Recently, Amoureux et al. have shown that efficient homonuclear dipolar proton decoupling can be achieved at high spinning speeds by applying smooth amplitude modulated (SAM) pulse sequences [4,44]. In this work, we used rotor-synchronized symmetry-based sequences [21,22] and PMLG schemes [5 9] during the evolution interval t 1 of MQC and SQC pulse sequences to reduce the 1 1 dipolar interactions, and consequently obtain line-narrowing in the indirect x 1 dimension of SQ and DQ 14 N spectra. We show that a substantially improved resolution can be achieved by merging SQC methods with symmetry-based sequences of the R-type. 2. Experimental Samples of L-alanine ( 14 N þ Ca Y a C b b COO ) and the tripeptide alanine alanine glycine (AAG, N þ Ca C CONC a C CON- C a 2 COO, 1 C labeled in all three C a positions) were purchased from Cambridge Isotope Laboratories. These samples were packed in a 2.5 mm outer diameter ZrO 2 rotor with a sample volume of ca. 11 ll. The spectra were obtained at 9.4 and 18.8 T, where the 1 ( 14 N) Larmor frequency appears at 400 (28.9) and 800 (57.8) Mz, respectively. The rotors were spun at 0.0 or 1.25 kz, either in a wide-bore Bruker triple resonance CP-MAS probe at 9.4 T (400 Mz for 1 ) or in a narrow-bore Bruker triple resonance CP-MAS probe at 18.8 T (800 Mz). In the former case, the magic angle was adjusted within 0.01 using deuterated a-oxalic[d 6 ] acid [45]; in the latter we used KBr to adjust the angle. The rf amplitude of the 14 N pulses was calibrated by direct detection of 14 N 4 NO, which has a very small quadrupole splitting. Using a 1 kw amplifier, the 14 N pulses had an amplitude of m N r = 57 kz. The details of the C, R and PMLG sequences are given below. 4. Recoupling heteronuclear interactions In all MQC and SQC experiments with L-alanine, we applied the R12 5 R12 5 rotor-synchronized pulse sequence [21,22] to protons S in the excitation and reconversion intervals s exc and s rec in order to recouple the heteronuclear dipolar S I interaction. We shall refer to R-MQC and R-SQC in these cases, where R indicates that we use symmetry-based sequences of the R-type to recouple heteronuclear dipolar interactions. This is shown in Fig. 1a and b. This windowless sequence consists of a sixfold repetition in rotor periods of the pulse pair { } 6, i.e., 12 pairs of p pulses with phases / =75 and 285. To ensure that the sequence is less sensitive to rf phase shift errors, phase-inverted supercycles [22,49] were used. Therefore the R12 5 element is followed by a R12 5 element where the signs of all phases are reversed. This supercycle reduces the number of second-order terms in the average amiltonian [46,49]. To further improve the performance of the recoupling of the heteronuclear dipolar S I interactions, nested supercycles may be implemented as shown by Brinkmann et al. for 17 O 1 pairs [25,26]. In experiments with AAG, the highest coherence transfer efficiency was obtained using the shorter R12 5 pulse sequence. 5. Decoupling homonuclear S S 0 interactions in the indirect dimension We have applied R6 1 and C6 2 pulse sequences to protons S in the evolution interval t 1 to decouple the homonuclear dipolar S S 0 interactions. If we opt for the R-MQC scheme in combination. Applications In L-alanine the 14 N nucleus in the rapidly rotating ammonium group only has a modest quadrupolar constant, which we have estimated [9] to be C Q = 1.1 Mz with g Q = In the 14 N þ group, the heteronuclear dipolar coupling between the three ammonium protons (which act together as spy nuclei) and the 14 N nucleus, when recoupled by the R sequence [46,47], is expected to be on the order of 1 2 kz. This interaction is partly overshadowed by the homonuclear S S 0 dipolar couplings between the ammonium 14 N þ protons, the nearby a, the more remote b protons and other protons belonging to neighboring molecules in the crystal lattice [27]. In the excitation and reconversion intervals, these interactions can be attenuated by appropriate R-type symmetry-based sequences, which simultaneously decouple the homonuclear S S 0 dipolar couplings while recoupling the heteronuclear S I interactions. omonuclear S S 0 couplings that act during t 1 can be suppressed by C-, R-type symmetry-based sequences or other homonuclear dipolar decoupling methods such as PMLG, DUMBO and SAM. In particular, PMLG and DUMBO perform best when the decoupling cycle time s c is not a multiple of the rotor period [41,42]. ere we have investigated the performance of a non-synchronized PMLG scheme. SAM is expected to be less efficient at high MAS frequencies [48]. Fig. 1. (a) R-MQC pulse sequence combined with homonuclear proton decoupling schemes for the indirect observation of 14 N SQ or DQ spectra. The rotorsynchronized R12 recoupling pulse scheme starts after a delay 1/(8s rot ) after the initial p/2 pulse. A repeating block with four p pulses as represented in (a) requires q = k =,6,12,... The dashed lines show the synchronization. For both the C6 2 and the R6 1 sequences, the first point of the 2D spectrum (i.e., n = 0) is obtained with t 1 = s rot =1/m rot, whereas the following points are obtained by incrementing t 1 = ndt 1 (with n =1,2,...) in steps Dt 1 =2/m rot for R6 1 and Dt 1 =4/m rot for C6 2. When C6 2 is used, m takes values 2(n 1), whereas for R6 1 m takes values (n 1), with n = 1, 2,... (b) R-SQC pulse sequence combined with homonuclear proton decoupling schemes for the indirect observation of 14 N SQ or DQ spectra. The excitation and reconversion conditions are similar to (a). During t 1, for both the C6 2 and the R6 1 sequences, the first point of the 2D spectrum (i.e., n = 0) is obtained with t 1 = s rot =1/m rot, whereas the other points are obtained by incrementing t 1 = ndt 1 (with n =1,2,...) in steps Dt 1 =1/m rot for R6 1 or Dt 1 =2/m rot for C6 2. When C6 2 is used, m takes values 2(n 1), whereas for R6 1, m takes values (n 1), with n = 1, 2,... The phase cycles are available from the authors upon request.

4 S. Cavadini et al. / Journal of Magnetic Resonance 202 (2010) with the R6 1 sequences, the latter must be inserted before and after the refocusing p pulse applied in the middle of the t 1 evolution period (Fig. 1a). Each R6 1 sequence consists of pairs of p pulses with phases / =90 and 270, i.e., six p pulses in one rotor period. On the other hand, the C6 2 windowless sequence consists of pairs of 2p pulses per rotor period with phases / = 0 and 180. Each C6 2 sequence needs two full rotor periods to be completed. In R-MQC experiments the advantage of R6 1 is thus that the smallest time increment required is Dt 1 =2s rot, i.e., half of what is needed for C6 2, which requires Dt 1 =4s rot. Consequently, the spectral width in the x 1 dimension can be twice as large for R6 1 as for C6 2. If we opt for the R-SQC scheme in combination with homonuclear decoupling schemes, the latter must be inserted between the two p/2 pulses applied to protons (Fig. 1b). In R-SQC experiments the smallest time increment is given by the number of rotational periods, e.g., Dt 1 = s rot for R6 1 and Dt 1 =2s rot for C Results and discussion Fig. 2 shows the 14 N SQ (top) and DQ (bottom) 2D R-MQC spectra of L-alanine ( 14 N þ CC COO ) obtained at 9.4 T. In addition to the strongest signal due to the ammonium protons of the N þ group, it is possible to observe weaker signals of other protons. The latter are weak compared to the ammonium protons because of the low efficiency of coherence transfer from remote 1 to 14 N and back. Long-range correlations may in principle be observed by increasing the excitation and reconversion intervals. The three ammonium protons, which rotate rapidly about the N C a axis, act together as three spy nuclei with S = 1/2 each, with a group spin S = /2. The efficiency of coherence transfer of the R-MQC and R-SQC methods can be estimated from the ratio r ¼ S SQ=DQ : ð1þ S spin-echo This ratio compares the signal amplitude S SQ/DQ of the first row of the 2D spectrum obtained with s exc = s rec, using phase-cycles appropriate for either 14 N SQ or DQ detection, and the spin echo signal amplitude S spin-echo of a 1D spectrum obtained without any 14 N pulses. For instance, with R12 5 R12 5 sequences during s exc = s rec =6s rot = ls (0.0 kz MAS), we have found that in L-alanine at 9.4 T the experimental efficiency of the two-way coherence transfer process is r = 0.17 and 0.08 for 14 N SQ and DQ, respectively. Without R12 5 R12 5 sequences, the RDS for a 1 14 N pair is merely on the order of 0.1 kz at 9.4 T, thus requiring s exc = s rec = 5 ms if there were no homogeneous T 0 2 decay. Empirically, the optimum intervals were found to be s exc = s rec = 1 ms if no recoupling is used [10]. Recoupling by R12 5 R12 5 sequences allows one to reduce s exc and s rec by a factor 5, thereby considerably reducing T 0 2 losses in these fixed intervals. The projections (rather than cross-sections) onto the x 1 axis are shown along the right-hand side of each 2D spectrum in Fig. 2. In the SQ spectra (top row), the 14 N þ group leads to a 14 N powder lineshape which spans about kz without resorting to any homo- Fig. 2. Experimental R-MQC spectra showing the 14 N SQ (a, b, c) and DQ (d, e, f) responses of the 14 N þ ammonium group of L-alanine (14 N þ CC COO ) with natural isotopic abundance, i.e., without replacing the remaining protons by deuterium nuclei. A 11 ll sample in a 2.5 mm rotor was spun at 0.0 kz (s rot =. ls) in a static field of 9.4 T (28.9 and 400 Mz for 14 N and 1 ). The rf field amplitudes were m 1 ( 1 ) p/2 = m 1 ( 1 ) p = 120 kz and m 1 ( 1 ) R12 = kz in the excitation and the refocusing intervals s exc = s rec = ls. Each R12 5 R12 5 sequence was comprised of 6 rotors periods. With m 1 ( 14 N) = 57 kz, the optimum 14 N pulse lengths were s p =18ls for SQ and s p =22ls for DQ. (a and d) The evolution period t 1 only contains a p S pulse applied to the protons. (b and e) Before and after the p pulse in the middle of the evolution period t 1 = ndt 1, two R6 1 sequences were inserted. (c and f) Likewise, two C6 2 sequences were inserted in the evolution period t 1. The simulations of the SQ spectrum (far right) were calculated for C Q = 1.1 Mz and g Q = 0.28, assuming uniform excitation of 4180 crystallites, generated with the Zaremba Conroy Wolfsberg (ZCW) algorithm [52 54]. The vertical x 1 axis is calibrated to 0 ppm for N 4 Cl (left side), which coincides with the 14 N carrier at 0 kz (right side). Each of the 2D spectra resulted from averaging 256 transients for each of 64 t 1 increments with Dt 1 =2/m rot = 66.6 ls for R6 1 sequences and Dt 1 =4/m rot = 1.2 ls for C6 2 sequences. The relaxation interval between consecutive scans was 4 s.

5 60 S. Cavadini et al. / Journal of Magnetic Resonance 202 (2010) 57 6 nuclear decoupling in the t 1 evolution period (Fig. 2a), which can be reduced to about 2 kz by inserting either C6 2 sequences (Fig. 2b) or R6 1 sequences (Fig. 2c) in the t 1 evolution interval. Similarly, for DQ spectra (bottom row), the projections onto the x 1 dimensions shown along the right-hand side of each 2D spectrum, reveal that the 14 N þ group leads to a signal which spans about 4 kz (Fig. 2d) that can be reduced to 2.5 kz by resorting either to C6 2 sequences (Fig. 2e) or to R6 1 sequences (Fig. 2f). The spectrum with the greatest similarity to the ideal simulated powder pattern was obtained with the R6 1 sequence (Fig. 2c and f). owever, the spectra obtained with such sequences (Fig. 2c and f) appear to be noisier than those obtained with C6 2 sequences (Fig. 2b and e). The difference might be due to the nature of the recoupling scheme. Indeed, the C6 2 sequence gives a scaling factor k = 0 for the isotropic chemical shift, whereas the R6 1 scales the isotropic chemical shift by a theoretical scaling factor k = 0.45 with an effective field along the x-axis [22,0,47]. This is represented by a symmetry-allowed first-order rotating-frame average amiltonian acting on the protons S during the evolution time t 1 ð1þ ¼ 0:45ðX S S x þ 2pJS x I z Þ; where X s represents the offset of the proton S with respect to the rf carrier, while J is the scalar coupling constant. In all these experiments, the C and R sequences are applied to the protons S when the density operator contains terms as S x I x, S y I x, S x I x 2 and S y I x 2. The success of these schemes is not limited to initial states with transverse proton magnetization such as in MQC-like experiments, but can be extended to initial states with longitudinal proton magnetization such as S z I x and S z I x 2 as occur in SQC-like experiments [50]. Indeed, the SQC method offers the possibility ð2þ Fig.. Experimental spectra showing the 14 N SQ responses of the 14 N þ ammonium group of L-alanine (14 N þ CC COO ). The 2.5 mm rotor was spun at 1.25 kz (s rot = 2 ls) in a static field of 9.4 T (28.9 and 400 Mz for 14 N and 1 ). The rf field amplitudes were m 1 ( 1 ) p/2 = m 1 ( 1 ) p = 120 kz and m 1 ( 1 ) R12 = 62.5 kz in the excitation and the refocusing intervals s exc = s rec = 192 ls. Each R12 5 R12 5 sequence was comprised of 6 rotor periods. With m 1 ( 14 N) = 57 kz, the optimum 14 N pulse length was s p =10lsfor SQ. (a) Experimental R-MQC spectra showing a linewidth of 2 kz in the indirect dimension. (b) Experimental R-SQC spectra with a linewidth of 1. kz showing a linenarrowing effect compared to the R-MQC spectra. The evolution period t 1 only contains a p pulse applied to the protons. (c) R-SQC spectra with R12 2 symmetry-based homonuclear decoupling during the evolution period t 1. (d) PMLG xx pp can be implemented in order to decouple homonuclear interactions during the evolution period t 1. The PMLG scheme comprises 12 pulses of 1 ls each, with an rf amplitude of 150 kz, and a delay of 2 ls between consecutive groups of 6 pulses. The cycle was applied twice, leading to a total duration of 2 ls, corresponding to one rotor period. Each of the 2D spectra resulted from averaging 128 transients for each of 64 t 1 increments. The vertical x 1 axis is calibrated to 0 ppm for N 4 Cl (left side), which coincides with the 14 N carrier at 0 kz. The relaxation interval between consecutive scans was s.

6 S. Cavadini et al. / Journal of Magnetic Resonance 202 (2010) of implementing rotor-synchronized pulse sequences with time increment given by Dt 1 = s rot, which is not achievable with the MQC scheme, since a p pulse has to be applied in the middle of t 1, between two rotor-synchronized pulse sequences. Therefore, the SQC method is preferable, since the spectral width in the indirect dimension is larger. In this work, we have investigated a series of decoupling schemes in combination with the R-SQC experiment. This is shown in Fig.. Depending on the homonuclear decoupling pulse sequences, different full widths at half maximum (FWM) have been measured. A remarkable improvement has been observed by comparing R-MQC with R-SQC. The FWM goes from 2 kz in R-MQC (Fig. a) to 1. kz in R-SQC (Fig. b). The latter can be further reduced to 1.2 kz (Fig. e) by applying homonuclear decoupling schemes such as the R12 2 during each t 1 increment of the R-SQC scheme. The PMLG xx pp scheme, repeated twice per rotor period, did not lead to any remarkable improvement of the resolution (Fig. d). We observed a strong dependence of the efficiency of the PMLG scheme upon the proton rf carrier frequency, so that a fairly good resolution could be obtained by this method by adjusting the rf carrier frequency. In Fig. 4 we plot the x 1 projections of the 2D spectra shown in Fig. with simulations of a two-spin system (I, S) under the effects of quadrupolar and other recoupled interactions. As in Fig., we have investigated the efficiency of C6 2 and R6 1 applied in combination with the R-SQC scheme. While the former does not perform well, the latter introduces an additional chemical shift term that distorts the 14 N spectra. The resulting patterns are plotted in Fig. 5 and can be explained by the recoupled first-order chemical-shift average amiltonian for 1 (spin S). Under the effect of the symmetry-based sequence R6 1, we found that the amiltonian can be described by ð1þ CS ¼ X S T S 11 XS T S 1 1 þ X S T S 10 ; ðþ (a) = 0.45 kz = 0.45 = 0 (b) = 2.4 kz = 0.7 = 0.18 (c) = 4.5 kz = 0.7 = 0.19 (d) = 6.1 kz = 0.5 = 0.19 (e) = 8.5 kz = 0. = 0.21 (a) R-MQC (kz) 5 10 (b) R-SQC (c) R-SQC + R12 2 Fig. 5. (a e) Projections onto the x 1 axis of experimental 2D proton-detected SQ 14 N spectra of L-alanine obtained at 18.8 T (800 Mz for 1 ) with 1.25 kz spinning frequency. The spectra were acquired with different 1 rf carriers, i.e., with different 1 offsets X Y. The scaling factors and refer to Eq. (). The continuous lines correspond to simulations (with 2.5 kz line-broadening) of 14 N SQ spectra obtained with C Q = 1.1 Mz and g Q = 0.28, assuming uniform excitation of 4180 crystallites, generated with the ZCW algorithm [52 54]. (d) -8 R-SQC + PMLG ω 1 /2π (kz) Fig. 4. (a d) Experimental proton-detected SQ 14 N spectra of L-alanine obtained by the projection onto the x 1 axis of Fig.. The dashed lines correspond to simulations (with a line-broadening of 1 kz) of 14 N SQ spectra obtained with C Q = 1.1 Mz and g Q = 0.28, and assuming uniform excitation of 4180 crystallites, generated with the ZCW algorithm [52 54]. where and are scaling factors. When the rf carrier is set on resonance (Fig. 5a), = 0.45 and = 0, as expected from the theory [21,22]. Far off resonance (Fig. 5b e), the symmetry-based pulse sequence may be less efficient and consequently the experimental is smaller than the theoretical value. Further off resonance, we have observed a weak peak half-way between the two more intense peaks (Fig. 5b e). This triplet -like feature can be attributed to small errors of the p pulses used during the R6 1 pulse sequence that lead to a scaled Zeeman term with 0. In the MQC scheme, the p pulse in the middle of the t 1 evolution period refocuses the effect of the above amiltonian acting on the protons S thus preventing the formation of the triplet (Fig. 2c and f). The R-SQC sequence was applied to the tripeptide AAG. Good resolution was achieved without applying any decoupling scheme

7 62 S. Cavadini et al. / Journal of Magnetic Resonance 202 (2010) 57 6 C Q = 2.6 Mz η Q = DQ R-SQC C Q = 1.1 Mz η Q = ω 1 /2π (kz) δ 2 (ppm) Fig. 6. Experimental and simulated (red continuous and dashed blue lines) R-SQC spectra showing the 14 N DQ responses of AAG (N þ Ca C CONC a C CON- C a 2 COO ). The 2.5 mm rotor was spun at 1.25 kz (s rot =2ls) in a static field of 18.8 T (57.8 and 800 Mz for 14 N and 1 ). The R12 5 pulse sequence was applied during rotors periods leading to excitation and refocusing intervals s exc = s rec =96ls. The rf field amplitudes were m 1 ( 1 ) p/2 = m 1 ( 1 ) p = 120 kz and m 1 ( 1 ) R12 = 62.5 kz in the excitation and the refocusing intervals. With m 1 ( 14 N) = 57 kz, the optimum 14 N pulse length was s p =20ls for SQ. The 2D spectra resulted from averaging 512 transients for each of 64 t 1 increments. The vertical x 1 axis is calibrated to 0 ppm for N 4 Cl (left side), which coincides with the 14 N carrier at 0 kz. The spectrum was acquired with an 14 N offset X N = kz. The relaxation interval between consecutive scans was s. The simulations (with a line-broadening of 1 kz) of 14 N SQ spectra obtained with C Q = 2.6 Mz and g Q = 0.7 for the amide 14 N in the peptide bond between the two alanine residues and C Q = 1.1 Mz and g Q = 0.28 for the 14 N of the terminal amino group N þ.we assumed uniform excitation of 4180 crystallites, generated with the ZCW algorithm [52 54]. during t 1. The presence of two amide 14 N sites in AAG, characterized by strong quadrupolar interactions (on the order of 2 Mz), leads to broad lines, making decoupling unnecessary. Since AAG experiences local dynamics on the 100 nanosecond (ns) timescale [51], we decided to detect the DQ 14 N spectrum only, which is not affected by motion. This is shown in Fig. 6. The good match between experiments and simulations at 18.8 T allows one to extract the quadrupolar coupling constant and the asymmetry parameter of two distinct 14 N sites (out of three) in AAG. The two detected 14 N peaks correspond to the 14 N of the N þ group (C Q = 1.1 Mz and g Q = 0.28) of the terminal L-alanine and to the N group (C Q = 2.6 Mz and g Q = 0.7) in the peptide bond between the two L-alanine residues. The third 14 N, which belongs to the alanine glycine linkage, did not appear in the 2D R-SQC spectra, possibly because it may have a larger quadrupolar coupling, which would require different conditions. 7. Conclusions It has been shown that the indirect detection of 14 N spectra via protons can benefit from decoupling of homonuclear proton proton dipolar interactions. The resulting line-narrowing allows one to obtain higher 14 N SQ and DQ resolution and to extract more reliable quadrupolar parameters in amino acids and peptides. The experiments proposed in this work should further benefit from the advent of very fast spinning probes with very high rf amplitudes, which should improve the performance of all decoupling sequences used in this work. In cases where the 14 N SQ powder pattern is affected by motions, SQC may provide more accurate measurements compared to MQC, since it reduces the broadening due to strong proton proton dipolar interactions. Acknowledgments This work has been supported by the Swiss National Science Foundation (FNRS Grant ) and the Swiss Commission for Technology and Innovation (CTI Grant PFIW-IW). References [1] W.G. Proctor, F.C. Yu, The dependence of a nuclear magnetic resonance frequency upon chemical compound, Phys. Rev. 77 (1950) 717. [2] R.E. Stark, R.A. aberkorn, R.G. Griffin, N-14 NMR determination of N bond lengths in solids, J. Chem. Phys. 68 (1978) [] R. Tycko, P.L. Stewart, S.J. Opella, Peptide plane orientations determined by fundamental and overtone N-14 NMR, J. Am. Chem. Soc. 108 (1986) [4].J. Jakobsen,. Bildsoe, J. Skibsted, T. Giavani, N-14 MAS NMR spectroscopy: the nitrate ion, J. Am. Chem. Soc. 12 (2001) [5] L. Müller, J. Am. Chem. Soc. 101 (1979) [6] D. Massiot, F. Fayon, B. Alonso, J. Trébosc, J.P. Amoureux, Chemical bonding differences evidenced from J-coupling in solid state NMR experiments involving quadrupolar nuclei, J. Magn. Reson. 164 (200) 160. [7] D. Iuga, C. Morais, Z. Gan, D.R. Neuville, L. Cormier, D. Massiot, NMR heteronuclear correlation between quadrupolar nuclei in solids, J. Am. Chem. Soc. 127 (2005) [8] Z. Gan, Measuring amide nitrogen quadrupolar coupling by high-resolution N- 14/C-1 NMR correlation under magic-angle spinning, J. Am. Chem. Soc. 128 (2006) [9] S. Cavadini, A. Lupulescu, S. Antonijevic, G. Bodenhausen, Nitrogen-14 NMR spectroscopy using residual dipolar splittings in solids, J. Am. Chem. Soc. 128 (2006) [10] S. Cavadini, S. Antonijevic, A. Lupulescu, G. Bodenhausen, Indirect detection of nitrogen-14 in solids via protons by nuclear magnetic resonance spectroscopy, J. Magn. Reson. 182 (2006) 168.

8 S. Cavadini et al. / Journal of Magnetic Resonance 202 (2010) [11] S. Cavadini, S. Antonijevic, A. Lupulescu, G. Bodenhausen, Indirect detection of nitrogen-14 in solid-state NMR spectroscopy, ChemPhysChem 8 (2007) 16. [12] Z. Gan, Rotary resonance echo double resonance for measuring heteronuclear dipolar coupling under MAS, J. Magn. Reson. 18 (2006) 25. [1] Z. Gan, 1C/14N heteronuclear multiple-quantum correlation with rotary resonance and REDOR dipolar recoupling, J. Magn. Reson. 184 (2006) 9. [14] Z. Gan, J.P. Amoureux, J. Trébosc, Proton-detected N-14 MAS NMR using homonuclear decoupled rotary resonance, Chem. Phys. Lett. 45 (2007) 16. [15] R.K. arris, A.C. Olivieri, Quadrupolar effects transferred to spin-1/2 magicangle spinning spectra of solids, Prog. NMR Spectrosc. 24 (1992) 45. [16] J. McManus, R. Kemp-arper, S. Wimperis, Second-order quadrupolar dipolar broadening in two-dimensional multiple-quantum MAS NMR, Chem. Phys. Lett. 11 (1999) 292. [17] M.. Levitt, T.G. Oas, R.G. Griffin, Rotary resonance recoupling in heteronuclear spin pair systems, Isr. J. Chem. 28 (1988) 271. [18] T.G. Oas, R.G. Griffin, M.. Levitt, Rotary resonance recoupling of dipolar interactions in solid-state nuclear magnetic-resonance spectroscopy, J. Chem. Phys. 89 (1988) 692. [19] Z.. Gan, D.M. Grant, Rotational resonance in a spin-lock field for solid-state NMR, Chem. Phys. Lett. 168 (1990) 04. [20] Z.. Gan, D.M. Grant, R.R. Ernst, NMR chemical shift anisotropy measurements by RF driven rotary resonance, Chem. Phys. Lett. 254 (1996) 49. [21] M. Carravetta, M. Eden, X. Zhao, A. Brinkmann, M.. Levitt, Symmetry principles for the design of radiofrequency pulse sequences in the nuclear magnetic resonance of rotating solids, Chem. Phys. Lett. 21 (2000) 205. [22] M.. Levitt, Encycl. Nucl. Magn. Reson. 9 (2002) 165. [2] S. Cavadini, A. Abraham, G. Bodenhausen, Proton-detected nitrogen-14 NMR by recoupling of heteronuclear dipolar interactions using symmetry-based sequences, Chem. Phys. Lett. 445 (2007) 1. [24] J.D. van Beek, R. Dupree, M.. Levitt, Symmetry-based recoupling of O-17-1 spin pairs in magic-angle spinning NMR, J. Magn. Reson. 179 (2006) 8. [25] A. Brinkmann, A.P.M. Kentgens, Sensitivity enhancement and heteronuclear distance measurements in biological O-17 solid-state NMR, J. Phys. Chem. B 110 (2006) [26] A. Brinkmann, A.P.M. Kentgens, Proton-selective O-17-1 distance measurements in fast magic-angle-spinning solid-state NMR spectroscopy for the determination of hydrogen bond lengths, J. Am. Chem. Soc. 128 (2006) [27] S.P. Brown, Probing proton proton proximities in the solid state, Prog. NMR Spectrosc. 50 (2007) 199. [28] B.C. Gerstein, C. Chow, R.G. Pembleton, R.C. Wilson, Utility of pulse nuclear magnetic-resonance in studying protons in coals, J. Phys. Chem. 81 (1977) 565. [29] S. afner,.w. Spiess, Multiple-pulse line narrowing under fast magic-angle spinning, J. Magn. Reson., Ser. A 121 (1996) 160. [0] P.K. Madhu, X. Zhao, M.. Levitt, igh-resolution -1 NMR in the solid state using symmetry-based pulse sequences, Chem. Phys. Lett. 46 (2001) 142. [1] S. Paul, R.S. Thakur, P.K. Madhu, -1 homonuclear dipolar decoupling at high magic-angle spinning frequencies with rotor-synchronised symmetry, Chem. Phys. Lett. 456 (2008) 25. [2] M. Lee, W.I. Goldburg, Nuclear-magnetic-resonance line narrowing by a rotating rf field, Phys. Rev. 140 (1965) A1261. [] M. Mehring, J.S. Waugh, Magic-angle NMR experiments in solids, Phys. Rev. B 5 (1972) 459. [4] M.. Levitt, A.C. Kolbert, A. Bielecki, D.J. Ruben, igh-resolution 1 NMR in solids with frequency-switched multiple-pulse sequences, Solid State Nucl. Magn. Reson. 2 (199) 151. [5] E. Vinogradov, P.K. Madhu, S. Vega, igh-resolution proton solid-state NMR spectroscopy by phase-modulated Lee Goldburg experiment, Chem. Phys. Lett. 14 (1999) 44. [6] E. Vinogradov, P.K. Madhu, S. Vega, Phase modulated Lee Goldburg magic angle spinning proton nuclear magnetic resonance experiments in the solid state: a bimodal Floquet theoretical treatment, J. Chem. Phys. 115 (2001) 898. [7] M. Leskes, P.K. Madhu, S. Vega, Proton line narrowing in solid-state nuclear magnetic resonance: new insights from windowed phase-modulated Lee Goldburg sequence, J. Chem. Phys. 125 (2006). [8] M. Leskes, P.K. Madhu, S. Vega, A broad-banded z-rotation windowed phasemodulated Lee Goldburg pulse sequence for -1 spectroscopy in solid-state NMR, Chem. Phys. Lett. 447 (2007) 70. [9] M. Leskes, P.K. Madhu, S. Vega, Supercycled homonuclear dipolar decoupling in solid-state NMR: toward cleaner -1 spectrum and higher spinning rates, J. Chem. Phys. 128 (2008). [40] D. Sakellariou, A. Lesage, P. odgkinson, L. Emsley, omonuclear dipolar decoupling in solid-state NMR using continuous phase modulation, Chem. Phys. Lett. 19 (2000) 25. [41] A. Lesage, D. Sakellariou, S. ediger, B. Elena, P. Charmont, S. Steuernagel, L. Emsley, Experimental aspects of proton NMR spectroscopy in solids using phase-modulated homonuclear dipolar decoupling, J. Magn. Reson. 16 (200) 105. [42] E. Salager, R.S. Stein, S. Steuernagel, A. Lesage, B. Elena, L. Emsley, Enhanced sensitivity in high-resolution -1 solid-state NMR spectroscopy with DUMBO dipolar decoupling under ultra-fast MAS, Chem. Phys. Lett. 469 (2009) 6. [4] J.P. Amoureux, B. u, J. Trébosc, Enhanced resolution in proton solid-state NMR with very-fast MAS experiments, J. Magn. Reson. 19 (2008) 05. [44] J.P. Amoureux, B.W. u, J. Trébosc, Q. Wang, O. Lafon, F. Deng, omonuclear dipolar decoupling schemes for fast MAS, Solid State Nucl. Magn. Reson. 5 (2009) 19. [45] S. Antonijevic, G. Bodenhausen, igh-resolution NMR spectroscopy in solids by truly magic-angle spinning, Angew. Chem. Int. Edit. 44 (2005) 295. [46] A. Brinkmann, M. Eden, Second order average amiltonian theory of symmetry-based pulse schemes in the nuclear magnetic resonance of rotating solids: application to triple-quantum dipolar recoupling, J. Chem. Phys. 120 (2004) [47] A. Brinkmann, M.. Levitt, Symmetry principles in the nuclear magnetic resonance of spinning solids: heteronuclear recoupling by generalized artmann ahn sequences, J. Chem. Phys. 115 (2001) 57. [48] M. Leskes, S. Steuernagel, D. Schneider, P.K. Madhu, S. Vega, omonuclear dipolar decoupling at magic-angle spinning frequencies up to 65 kz in solidstate nuclear magnetic resonance, Chem. Phys. Lett. 466 (2008) 95. [49] A. Brinkmann, J.S. auf der Günne, M.. Levitt, omonuclear zero-quantum recoupling in fast magic-angle spinning nuclear magnetic resonance, J. Magn. Reson. 156 (2002) 79. [50] S. Cavadini, A. Abraham, G. Bodenhausen, Coherence transfer between spy nuclei and nitrogen-14 in solids, J. Magn. Reson. 190 (2008) 160. [51] S. Cavadini, A. Abraham, S. Ulzega, G. Bodenhausen, Evidence for dynamics on a 100 ns time scale from single- and double-quantum nitrogen-14 NMR in solid peptides, J. Am. Chem. Soc. 10 (2008) [52] V.B. Cheng,.. Suzukawa, M. Wolfsberg, Investigations of a nonrandom numerical-method for multidimensional integration, J. Chem. Phys. 59 (197) 992. [5]. Conroy, Molecular Schrödinger equation. VIII. A new method for the evaluation of multidimensional integrals, J. Chem. Phys. 47 (1967) 507. [54] S.K. Zaremba, Good lattice points, discrepancy, and numerical integration, Ann. Mat. Pure Appl (1966) 25.